the lectern banner

By Steve Kelman

Blog archive

Evidence-based budgeting: Everything old is new again

FCW recently ran an article about a new OMB memo regarding the use of program effectiveness evidence in connection with agency submissions for the FY14 budget. The article topped the list of FCW most-read articles for a while, showing there is interest in this topic.
“Evidence-based government” is not a new idea. The “planning-programming budgeting system” of the 1960’s and the “zero-based budgeting” idea in the 1970’s were both outgrowths of the idea that budget allocations should be based on evidence about program effectiveness. (These decades also saw a number of very expensive experiments testing the impact of a number of anti-poverty interventions, many of which came up dry, showing the programs didn’t help very much against poverty.) The idea of “performance-based budgeting,” promoted by the Bush administration, was similar. Pretty much any time an administration proposes cutting back or eliminating a program, the proposal is based on evidence the program doesn’t work.
The basic idea behind evidence-based government was expressed in a comment attributed to John Maynard Keynes, to the effect that “when the facts change, I change my opinion – what do you do, sir?” And the basic problem with evidence-based government is that there is a lot of evidence for a psychological mechanism whereby most people don’t behave the way Keynes said he behaved: When contrary facts come in, most people who don’t like them don’t change their opinion. Instead they challenge the facts.

That is certainly a way of life in Washington, augmented by the fact that each side often has its own studies coming to opposite conclusions. Many in Washington are cynical about evidence.
This cynicism goes too far. Academics who are trained in research methods are constantly vetting academic studies, and there frequently emerge something close to a consensus about what conclusions are well-supported (or not) by evidence. Just calling something a “study” doesn’t make it so, and, if people in Washington were willing to place some trust in conclusions about different studies drawn by the scholarly community, the status of evidence for and against different programs, where it exists, would be better-established than the cynical story has it. 
Probably the best we can hope for, in Washington’s political world, is for evidence to be an input, and the scholarly community that seeks to develop evidence be an accepted participant, in policy debates. And that, I actually think, is the case now.
Since the idea of evidence-based government is not new, there is in a sense less to the OMB memo than meets the eye. Actually, its contributions are two. First, the memo actually is less about the use of evidence in backing up budget requests than it is about agencies using more resources, even in tight budget times, to gather evidence about programs. The memo rightly points out that new methods have been developed since the costly studies of the 1970’s to gather evidence about programs at much lower cost – including the overlap between performance information agencies gather for performance improvement purposes and the world of evidence-based government.

Second, ever since the often-negative results produced by the program evaluations of the 1970’s, Democrats – who otherwise often denounce Republicans for opposition to science and to reason – have often been uneasy about the idea of evidence-based government, fearing it will lead to program cuts. So it is good to see a Democratic administration sign on to this idea.

Posted on May 25, 2012 at 12:09 PM

FCW in Print

In the latest issue: grid security, agile development, BYOD and more!


Reader comments

Tue, May 29, 2012 SPMayor Summit Point, WV

Evidenced based budgeting -- reminds me in some ways of zero based budgeting. Planning and justifying a budget is a totally different matter from accounting for the subsequent spend. I expect this effort to be difficult for agencies - just as preparing performance based statements of work are.Evidence based budgets and PB SOWs require a major shift in how one determines what the needed inputs or funds are to achieve a set of outcomes. These outcomes need to be scalable to the level of funding provided, with a determination to be made at what point does the continued investment yield less of a return than justified or desired.One also needs to realize that as evidence is gathered and presented, and presumably publicly available, more than one jury of peers will come to different conclusions on the merits and accomplishments sought.

Source: FCW (

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above

More from 1105 Public Sector Media Group